Dixon’s Q test can be used to detect an outlier in a data set.
Usage:
(1) small data set
(2) single outlier suspected
Steps:
(1) Count the number of items in the data set.
(2) Sort the data from minimum to maximum.
(3) Determine the level of significance (alpha) to be used.
(4) Look up the Q value for the size of the data set and the level of significance.
r10 for the minimum value =
= ((second to lowest) – (minimum)) / ((maximum value) – (minimum value))
r10 for the maximum value =
= ((maximum value) – (second to highest)) / ((maximum value) – (minimum value))
Result
|
Interpretation
|
r10 for minimum value > Q
|
probable outlier
|
r10 for maximum value > Q
|
probable outlier
|